Affiliation:
1. Iskenderun Technical University
Abstract
Cyber-attacks are one of the most critical problems that seriously threaten society. Whereas there are various presentations and ways of carrying out cyber-attacks, numerous mechanisms and techniques exist to defend applications. Many malware creators have chosen the Android operating system as a target due to its popularity. Thousands of new malware samples, aiming to infect new devices daily, are trying to circumvent the security measures implemented by Android app stores. This study experiments with a multi-layer perceptron model for Android malware detection. This proposed system is based on static analysis techniques on Android. We analyzed popular machine learning algorithms with a total number of 129013 applications (5560 malicious and 123453 harmless software). We achieved higher malware-detection rates of 97.60% in the iterations.
Publisher
Islerya Medikal ve Bilisim Teknolojileri
Reference27 articles.
1. AV-TEST, The Independent IT-Security Institute. Security Report: Facts and Figures. Magdeburg, 2020.
2. Zhou Y, Jiang X. Dissecting android malware: Characterization and evolution. 2012 IEEE Symposium on Security and Privacy 2012; pp. 95-109.
3. Castillo CA. Android malware past, present, and future. White Paper of McAfee Mobile Security Working Group 1 2011; p. 16.
4. Andronio N, Zanero S, Maggi F. Heldroid: Dissecting and detecting mobile ransomware. International workshop on recent advances in intrusion detection, 2015; pp. 382-404.
5. Martinelli F, Mercaldo F, Nardone V, Santone A, Vaglini G. Model checking and machine learning techniques for HummingBad mobile malware detection and mitigation. Simulation Modelling Practice and Theory 2020; 105: 102169.